The ultimate objective of this proposal is to understand the genetic elements driving sporadic prostate cancer across multiple ethnic groups. Both somatic and, more recently, inherited genetic data highlight a 490 kilobase (kb) noncoding region on chromosome 8q24 as playing a major role in prostate cancer biology across multiple ethnic populations. At least 7 alleles are associated with prostate cancer risk. This finding represents the first validated genetic factor responsible for an appreciable amount of risk in the general population. Aim 1 intends to identify the actual causal alleles at 8q24 contributing to prostate cancer risk across five racial/ethnic populations. First, full characterization of common genetic variation in this 490 kilobase region will be determined by resequencing a multiethnic panel of 48 individuals. Second, all novel SNPs in this region will be genotyped in the well characterized HapMap samples to create a complete collection of all common genetic variation. Third, any newly discovered variants not adequately captured by our previous studies (as assessed by their correlations in HapMap) will be tested for association with prostate cancer in the MEC populations (2,788 incident prostate cancer cases and 2,613 controls). Aim 2 focuses on the intersection between inherited variation at 8q24 and the somatic phenotypes of gene expression and amplification. A total of 200 fresh frozen prostate tumor tissues will be analyzed (150 European American men and 50 African American men) for both projects. All of these samples will be genotyped for the known inherited risk alleles as well as any that are discovered during the course of this project. Since the risk alleles are noncoding, one hypothesis is that they are elevating risk by modulating expression levels of a gene in the vicinity. The expression study will take place in two stages. First, a comprehensive expression analysis covering 3.8 megabases of the 8q24 region will be assessed by tiling arrays to capture both annotated and unannotated transcribed sequences. Forty men representing the extremes of the risk allele distribution will be selected for this stage. Second, any candidate differentially expressed sequence will be validated in an independent sample of 160 men. Having identified a germline risk variant provides the unique opportunity to explore connections between the germline and somatic genomes. Amplification of the 8q region is one of the most frequent somatic lesions in prostate cancer. Tumors are often described as undergoing an evolutionary process of selection for tumor related traits. A new method based on this framework will be applied to evaluate if the risk allele resides on an amplified 8q chromosome more often than expected by chance. This observation would provide compelling evidence that the risk allele is selected for and, therefore, critical for tumor evolution. RELEVANCE TO PUBLIC HEALTH: Identifying the genetic factors underlying prostate cancer provides the opportunity to identify individuals at risk of developing disease as well as to lend insight into pathways that can be modulated for therapeutic benefit. Our proposal aims to pinpoint the causal changes in DNA sequence and the gene that it influences to better understand how this chromosomal region is responsible for an appreciable fraction of prostate cancer in the general population.